
# Joe Kelley - jkelley38@gatech.edu
# CS-6440-O01 - Mini-Project 2 
# Please pull the live copy from https://github.com/JoeKelleygt/Mini-Project-2
# All instructions are located within Mini-project 2.pdf

# Imports 
import pandas as pd
import numpy as np
import plotly.offline as pyo
import plotly.express as px
import plotly.graph_objects as go
import plotly.io as pio
from plotly.subplots import make_subplots


pyo.init_notebook_mode()


# Pneumonia Data from:
# https://catalog.data.gov/dataset/deaths-from-pneumonia-and-influenza-pi-and-all-deaths-by-state-and-region-national-center-
# Modified above to only include National data (not state breakdown) & only include all ages

df_pnaDeaths = pd.read_csv("../data/National_Pneumonia_Deaths.csv")

# Separate the MMWR report column to be more legible in a graph - by Week 
df_pnaDeaths['Week'] = df_pnaDeaths['MMWR Year/Week'].astype(str)
df_pnaDeaths['Week'] = df_pnaDeaths['Week'].str[-2:]

display(df_pnaDeaths)


# United States Historical Data - Pneumonia Deaths by Flu Season & Week
# Figure 1

fig_pnaPrev = px.scatter(df_pnaDeaths,x='Week', y='Deaths from pneumonia', color='season')
fig_pnaPrev.update_layout(title = "Yearly U.S. Pneumonia Deaths by Week - ICD-10 codes J12.0-J18.9")

display(fig_pnaPrev)


# Covid-19 Data from:
# https://data.cdc.gov/NCHS/Provisional-COVID-19-Death-Counts-by-Week-Ending-D/r8kw-7aab
df_covidUS = pd.read_csv("../data/Provisional_COVID-19_Death_Counts_by_Week_Ending_Date_and_State.csv")
df_covidUS["pna - pna_covid"] = df_covidUS["Pneumonia Deaths"] - df_covidUS["Pneumonia and COVID-19 Deaths"]

display(df_covidUS)


# Pneumonia Deaths in 2020 by State / All States
# Figure 2
fig_pna2020 = px.scatter(df_covidUS,x='End Week', y='Pneumonia Deaths', color='State')
fig_pna2020.update_layout(title = "2020 U.S. Pneumonia Deaths by Week - ICD-10 codes J12.0-J18.9")

display(fig_pna2020)


# Data taken from:
# https://www.cdc.gov/flu/weekly/

df_pnaCompare = pd.read_csv("../data/NCHSData37.csv")

display(df_pnaCompare)


# Comparing U.S. Pneumonia Deaths in 2020 vs Previous 6 Flu Seasons
# Figure 3

fig_pnaComparison = go.Figure(data = go.Scatter(x=df_pnaCompare['Week'], y=df_pnaCompare['Pneumonia Deaths'],
                                               mode='markers',marker=dict(
                                                   color=df_pnaCompare['Year'], colorscale='Sunset', showscale=True)))
fig_pnaComparison.update_layout(title = "U.S. Pneumonia Deaths 2020 vs Previous 6 Flu Seasons")

fig_pnaComparison.show()


# COVID-19 Deaths in 2020 by State / All States
# Figure 4
fig_covid2020 = px.scatter(df_covidUS,x='End Week', y='COVID-19 Deaths', color='State')
fig_covid2020.update_layout(title = "2020 U.S. Deaths involving COVID-19 - ICD-10 Code U07.1")

display(fig_covid2020)


# Deaths with Pneumonia and COVID-19 (ICD-10 codes J12.0-J18.9 and U07.1) in 2020 by State / All States
# Figure 5
fig_covid_pna2020 = px.scatter(df_covidUS,x='End Week', y='Pneumonia and COVID-19 Deaths', color='State')
fig_covid_pna2020.update_layout(title = "2020 U.S. Deaths with Pneumonia and COVID-19 - ICD-10 codes J12.0-J18.9 and U07.1")

display(fig_covid_pna2020)


# Influenza Deaths (ICD-10 codes J09-J11) in 2020 by State / All States
# Figure 6
fig_inf2020 = px.scatter(df_covidUS,x='End Week', y='Influenza Deaths', color='State')
fig_inf2020.update_layout(title = "2020 U.S. Influenza Deaths - ICD-10 codes J09-J11")

display(fig_inf2020)


# Deaths with Pneumonia, Influenza, or COVID-19 Deaths (ICD-10 codes J12.0-J18.9, J09-J11, U07.1) in 2020 by State / All States
# Figure 7
fig_covid_pna_inf_2020 = px.scatter(df_covidUS,x='End Week', y='Pneumonia, Influenza, or COVID-19 Deaths', color='State')
fig_covid_pna_inf_2020.update_layout(title = "2020 U.S. Deaths with Pneumonia, Influenza, or COVID-19 Deaths - ICD-10 codes J12.0-J18.9, J09-J11, U07.1")

display(fig_covid_pna_inf_2020)


# Comparison of the 4 figures above for reference
# Figure 8

subs1 = make_subplots(rows=1, cols=2, subplot_titles=('Pneumonia Deaths','COVID-19 Deaths'))
subs2 = make_subplots(rows=1, cols=2, subplot_titles=('Pneumonia and COVID-19 Deaths','Pneumonia, Influenza, or COVID-19 Deaths'))

subs1.add_trace(go.Scatter(x=df_covidUS['End Week'], y=df_covidUS['Pneumonia Deaths'], mode='markers',
                         marker=dict(color=df_covidUS['Pneumonia Deaths'],colorscale='deep')), row=1, col=1)

subs1.add_trace(go.Scatter(x=df_covidUS['End Week'], y=df_covidUS['COVID-19 Deaths'], mode='markers',
                         marker=dict(color=df_covidUS['COVID-19 Deaths'],colorscale='deep')), row=1, col=2)

subs2.add_trace(go.Scatter(x=df_covidUS['End Week'], y=df_covidUS['Pneumonia and COVID-19 Deaths'], mode='markers',
                         marker=dict(color=df_covidUS['Pneumonia and COVID-19 Deaths'],colorscale='deep')), row=1, col=1)

subs2.add_trace(go.Scatter(x=df_covidUS['End Week'], y=df_covidUS['Pneumonia, Influenza, or COVID-19 Deaths'], mode='markers',
                         marker=dict(color=df_covidUS['Pneumonia, Influenza, or COVID-19 Deaths'],colorscale='deep')), 
              row=1, col=2)

subs1.update_layout(showlegend=False)
subs2.update_layout(showlegend=False)

subs1.show()
subs2.show()


# Let's Subtract Pneumonia Deaths that were classified with the COVID-19 ICD-10 Code U07.1 from the Deaths classified
# as Pneumonia Deaths ICD-10 Codes J12.0-J18.9 to see how this looks
# Figure 9

fig_pna_noCovid = px.scatter(df_covidUS,x='End Week', y='pna - pna_covid', color='State')
fig_pna_noCovid.update_layout(title = "2020 U.S. Pneumonia Deaths Without COVID-19 ICD-10 Code U07.1")

display(fig_pna_noCovid)


# Let's combine some data - historical pneumonia deaths YoY for the United States vs Figure 9 (2020 PNA deaths without 
# ICD-10 Code U07.1)

# I decided to manually combine these data sets; while it can be done programmitcally, I experienced issues with my pandas
# data frames combinations, so for this comparison, a 2 minute excel copy was easier
df_pnaDeathsCombined = pd.read_csv("../data/National_Pneumonia_Deaths_Modified.csv")

# Separate the MMWR report column to be more legible in a graph - by Week 
df_pnaDeathsCombined['Week'] = df_pnaDeathsCombined['MMWR Year/Week'].astype(str)
df_pnaDeathsCombined['Week'] = df_pnaDeathsCombined['Week'].str[-2:]

display(df_pnaDeathsCombined)


# Let's see how the 2020 pneumonia data compares to historical trends 
# Figure 10

fig_pnaComb = px.scatter(df_pnaDeathsCombined,x='Week', y='Deaths from pneumonia', color='season')
fig_pnaComb.update_layout(title = "Annual U.S. Pneumonia Deaths by Week - ICD-10 codes J12.0-J18.9 NOT Including U07.1")
# print(fig_pnaComb.data)

fig_pnaComb.show()


# Let's take a look at the CDC's Nationally Notifiable Infectious Diseases and Conditions, United States: Weekly Tables -
# We'll see how cases during COVID-19's peak weeks compare to previous years trends
# https://wonder.cdc.gov/nndss/nndss_weekly_tables_menu.asp

# Hepatitis A
df_hepA = pd.read_csv("../data/infectious disease/hepA/hepA.csv")

display(df_hepA)


# Figure 11 - Hepatitis A

fig_hepA = px.bar(df_hepA,x='Week', y='Hepatitis Type A Cases', color='Year', barmode='group')
fig_hepA.update_layout(
    xaxis = dict(
    tickmode= 'linear',
    dtick = 1.0,
    ),
    title = "Hepatitis A Weekly Cases YoY (Weeks 13-18)")

display(fig_hepA)


# Infectious Diseases Data
# Chlamydia Trachomatis

df_ct = pd.read_csv("../data/infectious disease/chlamydia/chlamydia_data.csv")

display(df_ct)


# Figure 12 - Chlamydia Trachomatis

fig_ct = px.bar(df_ct,x='Week', y='Chlamydia Trachomatis Cases', color='Year', barmode='group')
fig_ct.update_layout(
    xaxis = dict(
    tickmode= 'linear',
    dtick = 1.0),
    title = "Chlamydia Trachomatis Weekly Cases YoY (Weeks 13-18)")

display(fig_ct)


# Infectious Diseases Data
# Giardiasis

df_gd = pd.read_csv("../data/infectious disease/giardiasis/giardiasis_cases.csv")

display(df_gd)


# Figure 13 - Giardiasis

fig_gd = px.bar(df_gd,x='Week', y='Giardiasis Cases', color='Year', barmode='group')
fig_gd.update_layout(
    xaxis = dict(
    tickmode= 'linear',
    dtick = 1.0),
    title = "Giardiasis Weekly Cases YoY (Weeks 13-18)")

display(fig_gd)

	geoid	Region	State	age	season	MMWR Year/Week	Deaths from influenza	Deaths from pneumonia	Deaths from pneumonia and influenza	All Deaths	Pecent of deaths due to pneumonia or influenza	pecent complete	Week
0	National	NaN	NaN	All	2017-18	201838	11	2808	2819	51480	5.475913	108.197858	38
1	National	NaN	NaN	All	2017-18	201815	212	3588	3800	55165	6.888426	115.942791	15
2	National	NaN	NaN	All	2012-13	201318	10	3504	3514	48368	7.265134	101.657226	18
3	National	NaN	NaN	All	2009-10	201027	3	3215	3218	45580	7.060114	95.797560	27
4	National	NaN	NaN	All	2010-11	201120	7	3575	3582	46635	7.680926	98.014901	20
...	...	...	...	...	...	...	...	...	...	...	...	...	...
489	National	NaN	NaN	All	2016-17	201640	18	3022	3040	51403	5.914052	108.036024	40
490	National	NaN	NaN	All	2016-17	201648	34	3515	3549	54594	6.500714	114.742694	48
491	National	NaN	NaN	All	2015-16	201601	32	4245	4277	55843	7.658972	117.367774	01
492	National	NaN	NaN	All	2015-16	201633	5	2866	2871	49809	5.764019	104.685842	33
493	National	NaN	NaN	All	2016-17	201645	19	3209	3228	52533	6.144709	110.410996	45

	Data as of	Start week	End Week	Group	State	Indicator	COVID-19 Deaths	Total Deaths	Percent of Expected Deaths	Pneumonia Deaths	Pneumonia and COVID-19 Deaths	Influenza Deaths	Pneumonia, Influenza, or COVID-19 Deaths	Footnote	pna - pna_covid
0	09/18/2020	02/01/2020	02/01/2020	By week	United States	Week-ending	0.0	58576.0	0.99	3796.0	0.0	479.0	4275.0	NaN	3796.0
1	09/18/2020	02/08/2020	02/08/2020	By week	United States	Week-ending	1.0	59296.0	0.99	3799.0	0.0	520.0	4320.0	NaN	3799.0
2	09/18/2020	02/15/2020	02/15/2020	By week	United States	Week-ending	0.0	58697.0	1.00	3824.0	0.0	558.0	4382.0	NaN	3824.0
3	09/18/2020	02/22/2020	02/22/2020	By week	United States	Week-ending	5.0	58734.0	1.01	3699.0	1.0	564.0	4267.0	NaN	3698.0
4	09/18/2020	02/29/2020	02/29/2020	By week	United States	Week-ending	9.0	59167.0	1.03	3822.0	5.0	654.0	4480.0	NaN	3817.0
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
1777	09/18/2020	08/15/2020	08/15/2020	By week	Puerto Rico	Week-ending	64.0	287.0	0.52	76.0	45.0	0.0	95.0	NaN	31.0
1778	09/18/2020	08/22/2020	08/22/2020	By week	Puerto Rico	Week-ending	45.0	193.0	0.37	38.0	32.0	0.0	51.0	NaN	6.0
1779	09/18/2020	08/29/2020	08/29/2020	By week	Puerto Rico	Week-ending	23.0	113.0	0.21	29.0	17.0	0.0	35.0	NaN	12.0
1780	09/18/2020	09/05/2020	09/05/2020	By week	Puerto Rico	Week-ending	36.0	69.0	0.13	27.0	23.0	NaN	40.0	One or more data cells have counts between 1–9...	4.0
1781	09/18/2020	09/12/2020	09/12/2020	By week	Puerto Rico	Week-ending	10.0	12.0	0.02	NaN	NaN	0.0	10.0	One or more data cells have counts between 1–9...	NaN

	Year	Week	Percent of Deaths Due to Pneumonia and Influenza	Expected	Threshold	All Deaths	Pneumonia Deaths	Influenza Deaths
0	2013	40	6.617957	6.36132	6.77011	47492	3140	3
1	2013	41	6.652714	6.45326	6.86185	47304	3135	12
2	2013	42	6.779127	6.55439	6.96279	47602	3216	11
3	2013	43	6.622544	6.66322	7.07142	47746	3151	11
4	2013	44	6.730631	6.77811	7.18613	48777	3271	12
...	...	...	...	...	...	...	...	...
358	2020	33	10.072165	5.13191	5.46758	58061	5843	5
359	2020	34	9.366954	5.12719	5.46286	54372	5085	8
360	2020	35	8.507472	5.13611	5.47178	48381	4106	10
361	2020	36	6.918100	5.15856	5.49423	36325	2509	4
362	2020	37	5.285226	5.19422	5.52989	18126	957	1

	geoid	Region	State	age	season	MMWR Year/Week	Deaths from influenza	Deaths from pneumonia	Deaths from pneumonia and influenza	All Deaths	Pecent of deaths due to pneumonia or influenza	pecent complete	Week
0	National	NaN	NaN	All	2009-10	201027	3.0	3215	3218.0	45580.0	7.060114	95.797560	27
1	National	NaN	NaN	All	2009-10	200952	47.0	4310	4357.0	49617.0	8.781264	104.282306	52
2	National	NaN	NaN	All	2009-10	200941	174.0	3703	3877.0	46453.0	8.346070	97.632384	41
3	National	NaN	NaN	All	2009-10	201032	5.0	2928	2933.0	44400.0	6.605856	93.317500	32
4	National	NaN	NaN	All	2009-10	201019	4.0	3464	3468.0	46362.0	7.480264	97.441125	19
...	...	...	...	...	...	...	...	...	...	...	...	...	...
522	NaN	NaN	NaN	NaN	2020	33	NaN	2595	NaN	NaN	NaN	NaN	33
523	NaN	NaN	NaN	NaN	2020	34	NaN	2434	NaN	NaN	NaN	NaN	34
524	NaN	NaN	NaN	NaN	2020	35	NaN	2134	NaN	NaN	NaN	NaN	35
525	NaN	NaN	NaN	NaN	2020	36	NaN	1711	NaN	NaN	NaN	NaN	36
526	NaN	NaN	NaN	NaN	2020	37	NaN	955	NaN	NaN	NaN	NaN	37

	Year	Week	Hepatitis Type A Cases
0	Y - 2016	13	14
1	Y - 2016	14	4
2	Y - 2016	15	11
3	Y - 2016	16	19
4	Y - 2016	17	18
5	Y - 2016	18	11
6	Y - 2017	13	17
7	Y - 2017	14	12
8	Y - 2017	15	13
9	Y - 2017	16	8
10	Y - 2017	17	8
11	Y - 2017	18	22
12	Y - 2018	13	52
13	Y - 2018	14	50
14	Y - 2018	15	47
15	Y - 2018	16	57
16	Y - 2018	17	81
17	Y - 2018	18	59
18	Y - 2019	13	190
19	Y - 2019	14	137
20	Y - 2019	15	131
21	Y - 2019	16	149
22	Y - 2019	17	192
23	Y - 2019	18	142
24	Y - 2020	13	18
25	Y - 2020	14	30
26	Y - 2020	15	28
27	Y - 2020	16	31
28	Y - 2020	17	38
29	Y - 2020	18	13

	Year	Week	Chlamydia Trachomatis Cases
0	Y - 2016	13	14552
1	Y - 2016	14	16154
2	Y - 2016	15	16220
3	Y - 2016	16	14126
4	Y - 2016	17	15551
5	Y - 2016	18	16725
6	Y - 2017	13	14739
7	Y - 2017	14	15810
8	Y - 2017	15	12684
9	Y - 2017	16	15778
10	Y - 2017	17	15756
11	Y - 2017	18	15819
12	Y - 2018	13	12158
13	Y - 2018	14	14268
14	Y - 2018	15	15402
15	Y - 2018	16	13094
16	Y - 2018	17	13263
17	Y - 2018	18	16466
18	Y - 2019	13	11054
19	Y - 2019	14	11493
20	Y - 2019	15	9459
21	Y - 2019	16	11129
22	Y - 2019	17	12711
23	Y - 2019	18	11859
24	Y - 2020	13	5896
25	Y - 2020	14	6883
26	Y - 2020	15	4884
27	Y - 2020	16	7085
28	Y - 2020	17	6090
29	Y - 2020	18	6371